Understanding my Dynamic balancer

[JMarkWolf]

I've used several dynamic balancers (Chadwick, DSS and others) on my two-bladed helicopter.

They all provide a velocity reading (ips) at a phase angle (zero to 359 degrees), relative to a opto-electronic trigger that starts (and presumably stops) the acquisition cycle when the "master" blade is directly "out front".

This reading is then manually plotted on a polar chart to find the corrective action.

A typical reading would be "0.5ips at 24 degrees".

Guys I've talked to that balance machinery for a living don't understand what the "24 degrees" refers to. They say that pure imbalances generate sine waves, and that maximum velocity always occur at 0 degrees and 180 degrees.

My question: If this is true, what does the 24 degrees in a typical readout "specifically" refer to? Is it the zero crossing "lag or lead" of the raw acceleration waveform?

 

[GregLocock]

I haven't used a rotor balancer, but it should be the same as a crankshaft or driveline balancer.

The angle is the phase lag, ie the location of the positive-going zero-crossing of the sinusoid of the velocity.

If in the example above you were to move the trigger further round by 24 degrees, then the readout would be 0.5 ips at 0 degrees.

Be very careful, acceleration, velocity and displacement sinusoids are at 90 degrees increment of phase lag.







Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[JMarkWolf]

So, the positive-going zero-crossing acceleration waveform is leading (or lagging) the "start" trigger by 24 degrees?

So then it wouldn't be possible to determine this phase relationship by capturing data only between the start and stop signals?

Sounds like a circular buffer would need to be employed to capture data prior to the trigger. Any suggestions as to where I might find a flow chart or psuedo code of this process?

 

[GregLocock]

Well, you want to average it in the time domain using a trigger from the start position, so one frame of data isn't much help.


I know how to build an expensive (FFT based) balancing rig, I don't know how to build a cheap one.



Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[JMarkWolf]

Greg Locock wrote:
"I know how to build an expensive (FFT based) balancing rig, I don't know how to build a cheap one"

I know of a balancing rig that has nothing more than a PIC microprocessor (possibly a DsPIC) and a 2-line LCD in it. The remote accelerometer, a MEMS type I suspect, and the photo-sensor connect directly to the A/D pin and digital input pin on the PIC. No signal conditioning at all!

It works pretty well, although I think it could benefit from some signal amplification, as the accelerometer signal diminishes as the balance become better.

It's not cheap either at $2200, considering the parts total up to less than $100.

Know of anyone I can commision to help me develope a similar unit?

 

[thruthefence]

Back in the "good old days" it was done with a flag, a grease pencil, a roll of masking tape and the pilots rear end. Believe it or not, I've worked a pilot that would say " give me four wraps of 2" tape on the red blade, and be damned if it was'nt < .2 IPS when checked with the vibrex! Can't get much cheaper then that. Kinda makes you pucker up a bit when that rotor disc tilts down towards you when you're first learning the technique.Thank goodness for technology!

 

[klyde]

Mark: Like you I'm a EE and own a two bladed helicopter. I have a Pro-Drive and a DSS balancer and could not always get a phase angle measurement that made sense. I made a bench set up with a D.C. motor and a flywheel that I could control the location of the center of mass of the flywheel. When I ran the flywheel at an rpm below it's critical speed, I could use the phase angle measurement to position the flywheel at that angle with respect to the optical sensor and then merely add corrective weight opposite the center of the shaft from the accelerometer. Works great and I always had good results. Until I ran the flywheel at an rpm above the critical speed and then there was a 180 degree shift in the phase angle information. By experimentation, I found that precisely at the critical speed, the phase angle was only 90 degrees off from the first run. This also applied to my helicopter. The main rotor I use "Add Opposite" for corrections. The tail rotor requires me to use "Add Same" for corrections. BTW we had an author describe the balancing without phase angle information and one of our subscribers balanced his tail rotor with just a dial indicator.